exercise central (aortic) blood pressure is predominantly ... · we conclude that elevation of...

Hughes and James E. SharmanMartin G. Schultz, Justin E. Davies, Phillip Roberts-Thomson, J. Andrew Black, Alun D.

Waves, Not Wave ReflectionNovelty and SignificanceExercise Central (Aortic) Blood Pressure Is Predominantly Driven by Forward Traveling

Print ISSN: 0194-911X. Online ISSN: 1524-4563 Copyright © 2013 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Hypertension doi: 10.1161/HYPERTENSIONAHA.111.00584

2013;62:175-182; originally published online May 28, 2013;Hypertension.

http://hyper.ahajournals.org/content/62/1/175World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://hyper.ahajournals.org/content/suppl/2013/05/28/HYPERTENSIONAHA.111.00584.DC1.htmlData Supplement (unedited) at:

http://hyper.ahajournals.org//subscriptions/

is online at: Hypertension Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialHypertensionin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on June 13, 2013http://hyper.ahajournals.org/Downloaded from

http://hyper.ahajournals.org/content/62/1/175

http://hyper.ahajournals.org/content/suppl/2013/05/28/HYPERTENSIONAHA.111.00584.DC1.html

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://hyper.ahajournals.org//subscriptions/

http://hyper.ahajournals.org/

175

A hypertensive response to exercise independently pre-dicts cardiovascular events and mortality.1 However, the

physiological mechanisms contributing to elevation in blood pressure (BP) with exercise are poorly understood. Previous studies have focussed on changes to brachial BP during exer-cise, but little attention has been directed toward understand-ing central (aortic) BP (the pressure to which organs such as the heart are exposed). Because of variations in pulse pres-sure amplification, central systolic BP is usually lower than brachial systolic BP and may differ greatly between individu-als with the same brachial systolic BP.2–5 Moreover, central to peripheral pressure amplification may be magnified during exercise,6,7 and we have also reported that individuals with increased cardiovascular risk because of hypercholesterol-emia have significantly raised central systolic BP and aug-mentation index, but not brachial systolic BP, during light to moderate exercise.8 Thus, risk related to BP may be better assessed by central BP during light to moderate intensity

exercise, rather than resting brachial BP. However, despite these observations, little is known of the contributing central hemodynamics.

At rest, with each heart beat a pressure wave generated by the left ventricle is propagated along the arteries and is reflected back toward the heart at sites of impedance mismatch. Return of this reflected pressure wave is proposed to be the sole contributor to augmentation (increase) in central BP.3 However, more recently it has been proposed that aortic pressure can be regarded as the sum of a reservoir pressure and an excess pressure, composed of discernibly discrete forward and backward propagating waves.9,10 Moreover, recent work has proposed that such discrete reflected waves make only a minor contribution to augmentation of the central BP under resting conditions.11 Whether the rise in central systolic BP with exercise can be attributed to increases in discrete forward or reflected waves, or the reservoir, has not been determined in humans. Given the large increase in systemic vasodilation that

Abstract—Exercise hypertension independently predicts cardiovascular mortality, although little is known about exercise central hemodynamics. This study aimed to determine the contribution of arterial wave travel and aortic reservoir characteristics to central blood pressure (BP) during exercise. We hypothesized that exercise central BP would be principally related to forward wave travel and aortic reservoir function. After routine diagnostic coronary angiography, invasive pressure and flow velocity were recorded in the ascending aorta via sensor-tipped intra-arterial wires in 10 participants (age, 55±10 years; 70% men) free of coronary artery disease with normal left ventricular function. Measures were recorded at baseline and during supine cycle ergometry. Using wave intensity analysis, dominant wave types throughout the cardiac cycle were identified (forward and backward, compression, and decompression), and aortic reservoir and excess pressure were calculated. Central systolic BP increased significantly with exercise (∆=19±12 mm Hg; P<0.001). This was associated with increases in systolic forward compression waves (∆=12×106±17×106 W·m−2·s−1; P=0.045) and forward decompression waves in late systole (∆=9×106±6×106 W·m−2·s−1; P<0.001). Despite significant augmentation in BP (∆=9±6 mm Hg; P=0.002), reflected waves did not increase in magnitude (∆=−1×106±3×106 W·m−2·s−1; P=0.2). Excess pressure rose significantly with exercise (∆=16±9 mm Hg; P<0.001), and reservoir pressure integral fell (∆=−5×105±5×105 Pa·s; P=0.010). Change in reflection coefficient negatively correlated with change in central systolic BP (r=−0.68; P=0.03). We conclude that elevation of exercise central BP is principally because of increases in aortic forward traveling waves generated by left ventricular ejection. These findings have relevance to understanding central BP waveform morphology and pathophysiology of exercise hypertension. (Hypertension. 2013;62:175-182.) • Online Data Supplement

Key Words: aorta ■ blood pressure ■ exercise ■ pulse wave analysis ■ venous reservoirs ■ wave intensity ■ wave reflection

Received November 5, 2012; first decision November, 27, 2012; revision accepted April 25, 2013.From the Menzies Research Institute Tasmania, University of Tasmania, Hobart, Australia (M.G.S., J.E.S.); International Centre for Circulatory Health,

National Heart and Lung Institute, Imperial College London, London, United Kingdom (J.E.D., A.D.H.); and Royal Hobart Hospital, Hobart, Australia (P.R.-T., J.A.B.).

The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA. 111.00584/-/DC1.

Correspondence to James E. Sharman, Menzies Research Institute Tasmania, University of Tasmania, Private Bag 23, Hobart 7000, Australia. E-mail [email protected]

Exercise Central (Aortic) Blood Pressure Is Predominantly Driven by Forward Traveling Waves, Not Wave ReflectionMartin G. Schultz, Justin E. Davies, Phillip Roberts-Thomson, J. Andrew Black, Alun D. Hughes,

James E. Sharman

© 2013 American Heart Association, Inc.

Hypertension is available at http://hyper.ahajournals.org DOI: 10.1161/HYPERTENSIONAHA.111.00584

Exercise and Central Blood Pressure


http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.111.00584/-/DC1



mailto:[email protected]


176 Hypertension July 2013

occurs during exercise,12 reduced augmentation of central BP during exercise might be expected. Therefore, we hypothesized that discernible wave reflection would be negligible with exercise and that the increase in central systolic BP would be predominantly driven by increased forward waves and possibly reduced aortic compliance. We aimed to determine this using wave intensity analysis and separation of invasively acquired aortic pressure into reservoir and wave components during exercise.

MethodsStudy ParticipantsTwenty patients scheduled to undergo routine diagnostic coronary angiography at the Royal Hobart Hospital, Hobart, Australia were recruited for participation in this study. All patients had a clinical indication for diagnostic angiography (eg, chest pain or positive exer-cise stress test) and were consented to participate before the clinical procedure, if there was a reasonable likelihood of meeting the study inclusion criteria. Inclusion criteria were a normal coronary angio-gram (<50% angiographic stenosis of any major coronary artery), preserved left ventricular function (ejection fraction >50%), and no aortic valve disease (<20 mm Hg aortic valve gradient). None of the participants had a clinical history of impaired sympathetic func-tion. On the basis of our previous invasive exercise study,13 in which there was a 26±14 mm Hg increase in aortic systolic BP from rest to exercise, we determined that 10 participants were required to com-plete the study to detect a ≥11 mm Hg change in aortic systolic BP (α=0.05, β=0.20; ie, less than half the expected systolic BP change from rest to exercise seen previously). A total of 20 patients were recruited to allow for data loss because of failure to meet inclusion criteria or other reasons. Of the 20 patients who consented, 6 were identified as having clinically significant coronary artery disease and were excluded. A further 2 patients experienced radial artery spasm during the catheterization procedure, and it was deemed inappropri-ate to continue. From the remaining 12 participants, data from 2 indi-viduals were excluded from the analysis because of poor-quality flow velocity recordings.

Study ProtocolAfter the standard coronary angiogram procedure (and confirmation of nonsignificant coronary vessel disease), hemodynamic measure-ment of ascending aortic pressure and flow velocity was made under baseline (resting) conditions and during supine cycle ergometry. Continuous measurements of hemodynamic data were made for ≤2 minutes during both resting and exercise conditions. Exercise was performed in the supine posture using a cycle ergometer (Monarch Rehab Trainer 881E, Sweden) which was placed on the catheter laboratory table. Load was set at ≈50 W, and patients were asked to cycle at a comfortable pace so that steady state heart rate at a light to moderate workload could be achieved, at which point all exer-cise hemodynamic measures were undertaken. Participant clinical characteristics were extracted from medical records. The study was approved by the University of Tasmania Human Research Ethics Committee, all participants provided written informed consent, and procedures were performed in accordance with the declaration of Helsinki.

Hemodynamic DataContinuous simultaneous pressure and flow velocity were recorded by a single-use 0.014″, straight tip intra-arterial pressure, and Doppler flow wire (Combowire, Volcano Therapeutic Corp, Rancho Cordova, CA). Arterial access was made via standard radial puncture, and measurements were performed through a 100-cm 5F rim cath-eter (Cook Medical, Bloomington, IL). The catheter was positioned in the ascending aorta ≈5 cm distal to the valve leaflets and turned 180° with the tip facing away from the valve and flow direction. The Combowire was advanced ≈5 mm beyond the catheter tip, where it

was stabilized by the catheter, away from the arterial wall and in the center of flow (confirmed by fluoroscopy). To obtain quality pres-sure and flow velocity, small rotational movements of the Combowire were made once it was positioned in the ascending aorta. Analogue outputs of the pressure and flow velocity signals were digitized (PowerLab ML870 8/30, AD Instruments, Bella Vista, Australia) and recorded using LabChart 7 software (AD Instruments). Data were acquired at a sampling rate of 1000 Hz, and a simultaneous 3-lead ECG recording was made to calculate heart rate. Pressure and flow velocity were calibrated offline using a 2-point calibration method. To obtain calibration values, 2 raw test signals of a known value were sent from the Combowire to acquisition software before measure-ment of hemodynamic data.

Aortic pressure and flow velocity captured during rest and exercise were ensemble averaged offline for ≤6 heart cycles. Aortic systolic BP was considered the maximum pressure point and the diastolic BP the minimum pressure point on the waveform. Aortic pulse pressure was defined as the difference between the systolic and diastolic BP. Aortic mean arterial pressure was calculated from integration of the waveform using custom written software. Aortic augmentation pres-sure (systolic BP−pressure at the first systolic shoulder or inflection point [P1]) was calculated using customized Matlab software to iden-tify the inflection point on the aortic pressure wave (taken as P1) and calculated augmentation pressure in relation to maximal (systolic) pressure. Augmentation pressure as a percentage of the overall pulse pressure was calculated to derive the augmentation index (AIx). Aortic wave speed was calculated using the sum of squares method at the measurement site as previously described.14 Reproducibility of pressure and flow velocity measures was good (see online-only Data Supplement).

Wave Intensity AnalysisIdentification of waves responsible for directing the flow of blood in the arterial segment was performed by wave intensity analysis. Waves were classified with respect to (1) their direction in relation to the direction of aortic flow (forward and backward) and the gra-dient of pressure change across the wavefront: waves with positive pressure gradients being termed compression waves, and waves with negative pressure gradient termed decompression waves. For example, in the proximal aorta, waves can originate from a proximal source (ie, the left ventricle) or from a distal source (eg, reflections from the peripheral vasculature). Acceleration of blood can result from compression waves of proximal origin or from decompres-sion waves of distal origin, whereas deceleration of blood can result from compression waves of distal origin or decompression waves of proximal origin. Previous studies in the aorta have identified a consistent pattern of 3 dominant waves as shown in Table 1.15,16 In some recordings smaller waves were evident (typically with inten-sities <5% of the intensity of the initial compression wave) but not analyzed.

Wave intensity data for each individual, at rest and during ex-ercise, were calculated from the ensemble averaged pressure and flow velocity data as previously described,17 and outlined in the online-only Data Supplement. Separation of pressure into forward and backward components was performed without (prereservoir subtraction) and after subtraction of the aortic reservoir pressure (postreservoir subtraction). Reflection coefficient was calculated as the ratio of peak backward to forward propagating pressure and multiplied by 100 to obtain percentage values. A wave reflection index was also calculated to determine the magnitude of wave re-flection, using the ratio of backward compression wave integral to forward compression wave integral, and multiplied by 100 to obtain percentage values. Impedance analysis was also performed as described by Hughes and Parker,17 and is also described in the online-only Data Supplement.

Reservoir and Excess Pressure SeparationEnsemble averaged aortic pressure was separated into 2 components (reservoir and excess pressure; Figure 1). Reservoir pressure can be calculated using both pressure and flow velocity, or from pressure



Schultz et al Exercise Central Hemodynamics 177

alone. Both methods produce quantitatively similar results, with only slight variation in shape of the initial upward inflection of reservoir pressure. As this did not alter the overall results (ie, the change in reservoir pressure from rest to exercise), and for simplicity, we present results from the pressure-only separation. This method of separation is outlined in the online-only Data Supplement. Excess pressure was calculated by subtracting the reservoir pressure from aortic pressure and represents pressure attributable to discrete forward and backward waves. Separated pressures are presented as peak or integral values.

Statistical AnalysisAll statistical calculations were made using PASW 18.0 (SPSS Inc, Chicago, IL). Data were visually inspected for normality of distribu-tion, and skewed data were log transformed if required. The change in all hemodynamic, wave intensity, and separated pressure wave parameters from rest to exercise were compared using paired Student t tests. Because of the shortening of the cardiac cycle, which occurs with exercise (increased heart rate), all integral data were addition-ally adjusted for heart rate, by dividing the integral values of wave intensity, reservoir, and excess pressure with heart rate (beats per minute). Pearson correlation coefficients (r) were calculated to assess relationships among hemodynamic, wave intensity, and wave sepa-ration variables at rest and during exercise. P<0.05 was considered statistically significant.

ResultsPatient CharacteristicsThe clinical characteristics of study participants are outlined in Table 2. Participants were predominantly middle aged, men, and of elevated body mass. Most participants received treatment for hypertension and hyperlipidemia. Two participants had type 2 diabetes mellitus and 4 had a smoking history (current or former).

Hemodynamic Parameters: Changes From Rest to ExerciseCentral systolic BP, mean arterial pressure, pulse pressure, P1, augmentation pressure, Aix, and heart rate increased

significantly from rest to exercise (P<0.05 for all; Table 3). Aortic wave speed also increased significantly from rest to exercise, indicating a decrease in proximal aortic compliance. There were no significant changes in central diastolic BP and aortic flow velocity.

Wave Intensity: Changes From Rest to ExerciseWave intensity data are presented in Table 4. Exercise caused significant increases in both the magnitude of the net forward compression and decompression waves. These remained significant after adjustment for heart rate for the forward compression waves (Δ=+19±24×104 W·m−2·s−1; P=0.03) and for the forward decompression waves (Δ=+14±19×104 W·m−2·s−1; P=0.001). The magni-tude of the backward compression (reflected) wave was not significantly changed, before or after adjustment for heart rate (Δ=+2±5×104 W·m−2·s−1; P=0.2). The reflection coef-ficient decreased from rest to exercise, whether analyzed as peak (Table 4) or integral data (−9±28%; P=0.4) but these changes were not statistically significant. Figure 2 shows a representative example of wave intensity analysis of the response to exercise. In this example, from rest to exercise, heart rate and aortic systolic pressure increased signifi-cantly, along with large increases in both forward compres-sion and decompression waves. Despite this, the backward compression (reflected) wave remained largely unchanged in magnitude.

Impedance and Forward and Backward Pressure (Without Reservoir Pressure Subtraction): Changes From Rest to ExercisePressure separation using impedance analysis was quantitatively similar to time-domain analysis, in that forward pressure increased significantly from rest to exercise, whereas backward pressure was not significantly changed (Table 4). Figure S1 in the online-only Data Supplement provides an example of wave separation at rest and during exercise using time-domain and frequency-domain approaches—the results are qualitatively indistinguishable. Analysis of impedance spectra plots was consistent with the findings of wave intensity analysis showing no evidence of increased wave reflection after exercise (Figure S2).

Forward and Backward Pressure (With Reservoir Pressure Subtraction): Changes From Rest to ExerciseFindings after subtraction of the aortic reservoir pressure were qualitatively similar to that of prereservoir extraction. Separated forward pressure increased significantly from rest to exercise and backward pressure remained unchanged. The reflection coefficient tended to be reduced from rest to

Table 1. Origin and Nature of Waves in the Ascending Aorta as Defined by Wave Intensity Analysis

Pressure Flow Velocity Wave Nature Wave Origin Proposed Cause of Wave

Increase Increase Forward compression Proximal Left ventricular ejection

Increase Decrease Backward compression Distal Reflection of forward compression wave

Decrease Decrease Forward decompression Proximal Deceleration of left ventricular ejection before aortic valve closure

Figure 1. Example of a central pressure waveform separated into its components of reservoir pressure (white shading) and the excess or wave pressure (gray shading).




exercise (Table 4), but these changes did not reach statistical significance.

Reservoir and Excess Pressure: Changes From Rest to ExerciseThe integral of the excess pressure increased significantly from rest to exercise, before (Table 4) or after accounting for heart rate (Δ=+4±4×103 Pa·s; P=0.008). Peak excess pressure increased significantly with exercise. From rest to exercise, the integral of the aortic reservoir pressure was significantly reduced (Table 4) and remained so after correcting for heart rate (Δ=−13±12×103 Pa·s; P=0.008). Peak aortic reservoir pressure was unchanged with exercise. At rest, reservoir pres-sure comprised 68% of total pressure, and during exercise it dropped to 49% of total pressure. A typical example of the pressure waveform separated into forward and backward

components at rest and during exercise in a 41-year-old male is depicted in Figure 3.

Associations Among Hemodynamic, Wave Intensity, and Separated Pressure VariablesUnder both rest and exercise conditions, aortic systolic BP correlated closely with peak reservoir pressure (r=0.810; P=0.005 and r=0.679; P=0.031, respectively). The change in central systolic BP was associated with the change in peak reservoir pressure (r=0.873; P=0.001). The change in aortic wave speed was associated with the change in backward compression and forward decompression waves (prereservoir subtraction: r=0.634; P=0.049 and r=0.681; P=0.030, respectively), as well as the change in the excess pressure integral (r=0.645; P=0.044). The change in the excess peak and excess integral pressure was strongly associated with the change in the integrated intensity of the forward compression wave (postreservoir subtraction: r=0.903; P<0.001 and r=0.633; P=0.050, respectively), but not the backward compression wave (postreservoir subtraction: r=0.224; P=0.535 and r=0.419; P=0.228, respectively). The change in the forward compression wave also positively correlated with the change in heart rate (postreservoir subtraction: r=0.672; P=0.033). There was also a significant negative (inverse) relationship between change in reflection coefficient (prereservoir subtraction integral values) and the change in central systolic BP (r=−0.682; P=0.030).

DiscussionThis is the first study to demonstrate that elevation in exer-cise central BP is accounted for by major increases to forward propagating pressure waves, and not wave reflection. This find-ing is contrary to some previous studies which, on the basis of the elevated AIx, have attributed pressure augmentation dur-ing exercise to increases in wave reflection. This finding is of relevance to understanding central BP hemodynamic mecha-nisms and may have important clinical ramifications given the independent prognostic value of a hypertensive response to exercise for predicting adverse cardiovascular outcomes.1

Augmentation of central BP is widely believed to be mediated via wave reflection. With each contraction of the heart, a compression wave associated with a rise in pressure and increased flow is propagated forward through the large

Table 2. Clinical Characteristics of the Study Population (n=10)

Variable Mean±SD or n (%)

Sex (men, %) 7 (70)

Age, y 55±10

Height, cm 173±11

Weight, kg 93±17

Brachial systolic blood pressure, mm Hg 126±15

Brachial diastolic blood pressure, mm Hg 79±11

Body mass index, kg·m−2 32±5

Treated hypertension 6 (60)

Antihypertensive medication

Angiotensin-converting enzyme inhibitor 2 (20)

Angiotensin receptor blocker 1(10)

Non-dihydropyridine calcium channel blocker 1 (10)

Diuretic 1 (10)

β-Blocker 1 (10)

Type 2 diabetes mellitus, % 2 (20)

Smoking history, % 4 (40)

Statin therapy, % 6 (60)

Smoking history is current or former. Blood pressure recorded at preprocedure examination.

Table 3. Aortic Hemodynamic Parameters at Rest and During Exercise (n=10)

Variable Rest Exercise Change P Value

Aortic systolic pressure, mm Hg 121±8 140±11 19±12 <0.001

Aortic diastolic pressure, mm Hg 74±7 80±14 6±11 0.13

Mean arterial pressure, mm Hg 90±5 100±11 10±11 0.015

Aortic pulse pressure, mm Hg 47±11 60±13 14±6 <0.001

P1, mm Hg 110±8 120±11 11±10 0.006

Augmentation pressure, mm Hg 12±8 21±10 9±6 0.002

Augmentation index, % 23±14 33±14 10±11 0.023

Heart rate, bpm 64±10 80±11 16±6 <0.001

Peak aortic flow velocity, cm·s−1 73±18 80±15 7±12 0.100

Aortic wave speed, m·s−1 6.5±2.2 7.9±2.2 1.4±1.1 0.004

Data are mean±SD.




arteries. When this incident wave reaches sites of impedance mismatch some of its energy will be reflected back toward the heart, augmenting central BP.3 In this study, using wave intensity analysis, we assessed the contribution of both forward and reflected waves to central BP augmentation during exercise. Augmentation of exercise BP occurred almost exclusively by major increases in forward propagating waves originating from the left ventricle, and not wave reflection, which was essentially unchanged during exercise. This finding supports recent observations that forward wave propagation and aortic reservoir characteristics (aortic compliance) may play a more dominant role in determining central BP.9,11,18–20

Exercise Hemodynamics: Forward and Reflected Waves in the AortaExercise triggers several acute hemodynamic changes. Under normal circumstances, systemic vasodilation offsets the rise in cardiac contractility, heart rate, and left ventricular output, resulting in increased peripheral blood flow.21 Arterial pressure, both peripheral and central, rises in a graded fashion with increasing exercise intensity,22 but less is known about the underlying hemodynamic mechanisms of this increase. Murgo et al23 measured invasive aortic pressure and flow velocity during exercise. They observed an exercise-induced increase in central systolic BP, coupled with a reduction in aortic compliance (as determined by an increase in aortic pulse wave velocity) and decreased peripheral vascular resistance. Although they reported a decrease in reflected wave transit time, the physiological stress of exercise was not sufficient to significantly alter the site or magnitude of wave reflections in the aorta. This observation is consistent with our finding of no appreciable changes in backward compression

waves or reflection coefficient with exercise. In our studies, the magnitude of reflection at rest and during exercise ranged between 2% and 20%, and changes caused by exercise were on average around −6 to −3% (ie, reduction). Such minimal change in wave reflection, despite profound vasodilation on exercise, might be explained by poor retrograde transmission of reflected waves in the periphery resulting in a horizon effect on wave travel.20 This suggests that entrapment and dissipation of reflected pressure waves occurs within the periphery and that there is minimal backward transmission of discrete reflected waves.19,20 This notion is in keeping with our findings, where reflected wave components in the aorta seemed relatively unaffected by exercise-induced peripheral vasodilation, despite large increases in the magnitude of the incident pressure wave. Interestingly, Laskey et al24 also found a reduction in the first harmonic of the reflection coefficient (suggesting diminished reflected wave magnitude) on initiation of exercise in healthy subjects. Adding to this observation, this current study found a strong and negative association between the change in magnitude of reflection and change in central systolic BP with exercise. Such results suggest that the cardiovascular system may be relatively close to an optimal design to maximize flow output by minimizing impedance to forward propagating waves,25 and that discrete reflections contribute only modestly to the morphology of the central BP waveform.

Contribution of Aortic Reservoir and Excess Pressure to Exercise Central BPAccounting for the systolic radial expansion and diastolic recoil of the arteries in the separation of the pressure wave-form into forward and backward components is the funda-mental precept of the reservoir-excess pressure paradigm.9

Table 4. Wave Intensity Analysis and Separated Pressure Wave Parameters at Rest and During Exercise (n=10)

Variable Rest Exercise Change P Value

Prereservoir subtraction

Forward compression wave, W·m−2·s−1 40×106±18×106 52×106±22×106 12×106±17×106 0.045

Backward compression wave, W·m−2·s−1 −7×106±4×106 −8×106±5×106 −1×106±3×106 0.2

Forward decompression wave, W·m−2·s−1 11×106±5×106 20×106±6×106 9×106±6×106 <0.001

Peak forward pressure, mm Hg 69±16 79±14 10±9 0.009

Peak backward pressure, mm Hg 12±7 9±5 −3±6 0.2

Reflection coefficient, % 18±12 12±7 −6±9 0.06

Wave reflection index, % 20±11 17±10 −3±10 0.3

Postreservoir subtraction

Peak forward pressure, mm Hg 61±15 74±14 13±9 0.009

Peak backward pressure, mm Hg 3±4 2±1 −2±4 0.2

Reflection coefficient, % 6±7 2±1 −4±7 0.2

Reservoir/excess pressure separation

Peak aortic reservoir pressure, mm Hg 110±5 114±17 4±16 0.5

Integral aortic reservoir pressure, Pa·s 18×105±7×105 13×105±5×105 −5×105±5×105 0.010

Peak aortic excess pressure, mm Hg 11±5 27±12 16±9 <0.001

Integral aortic excess pressure, Pa·s 9×105±3×105 13×105±4×105 5×105±3×105 <0.001

Data are mean±SD. Wave intensity values are net integrals (area under the curve). A negative change in backward waves indicates greater intensity. Peak forward and backward pressures are above diastolic pressure. Reflection coefficient was calculated using the ratio peak backward/peak forward pressure. Wave reflection index was calculated using the ratio of backward compression wave integral/forward compression wave integral. Data presented are not adjusted for heart rate.




By definition, the aortic reservoir pressure is the result of both flow into and out of the proximal aorta and is dependent on the interaction between aortic compliance and downstream impedance. It, therefore, has been suggested to represent the theoretical minimum work that the left ventricle must perform to force blood into the proximal aorta. The excess pressure is, therefore, an indicator of the excess work required of the left ventricle above this minimum.25 The reservoir-excess pres-sure paradigm has never been assessed in humans in the con-text of exercise, but provides a novel way to describe changes to exercise central BP in the current study. With exercise, we observed an increase in central systolic BP and heart rate. Concurrently, there may be marked sympathetic activation to diminish blood flow to nonessential vascular beds, while skeletal muscle activity causes vasodilation of peripheral

arterioles. Consequently, to meet the increased demand for blood flow during exercise, the heart must increase the excess pressure. The vasodilation of the peripheral vasculature and poor retrograde transmission would ensure the magnitude of wave reflections seen in the aorta is minor, and central BP is, therefore, augmented almost exclusively by incident pres-sure waves of myocardial origin. Indeed, in support of this view, we found that the change in both peak and integral of the excess pressure was positively associated with the change in the forward compression wave, but not reflected wave components.

Recent ex vivo modeled data criticized the reservoir-excess pressure paradigm, suggesting it introduces error

Figure 2. Wave intensity analysis (WIA, top) from aortic pressure (solid line) and flow velocity (broken line; bottom) measured in a 42-year-old male at (A) rest and (B) during exercise. From rest to exercise, heart rate and peak systolic aortic pressure increased significantly (67–85 bpm and 114–135 mm Hg, respectively). The direction of change in WIA from rest to exercise is outlined with arrows in B (↔, no change; ↑, increase). Both forward compression (FCW) and forward decompression (FDW) waves increased significantly with exercise, whereas the backward compression wave (BCW) did not change.

Figure 3. A typical example of ascending aortic pressure separated into the components of reservoir pressure (white shading) and excess pressure (gray shading) during rest (A) and exercise (B). In this 41-year-old male patient, from rest to exercise, heart rate and peak systolic pressure increased significantly (74–85 bpm and 129–143 mm Hg, respectively). There was no change in peak reservoir pressure; however, there was a large increase in excess pressure.




(overestimation of late systolic forward decompression waves) into hemodynamic analysis.26 It is important to note that the same pattern of wave intensity (large, significant increase to incident waves, and no change in reflected waves) emerged under both resting and exercise conditions. Irrespective of analysis technique, our data indicate only a minor role for wave reflection. Taken altogether, it seems plausible that the aortic reservoir-excess pressure paradigm provides a reason-able and physiologically sound explanation for the observed central BP changes with exercise that cannot be explained in terms of wave reflection alone.

LimitationsStudy participants were older individuals with an indication for coronary angiography and, whilst all were free of sig-nificant coronary vessel disease, they cannot be considered truly healthy. Therefore, results may not be generalizable to healthy or younger individuals where augmentation pres-sure and augmentation index estimated by radial tonometry may reduce with exercise.7,8,12,22 This study was adequately powered on the basis of expected changes in central systolic BP from rest to exercise. However, larger variance in other hemodynamic factors may have increased the probability of a type 2 error because of a small study sample (despite the large hemodynamic perturbations induced with exercise and the consistency in the direction of responses). In addition, a mathematical model was used to derive aortic reservoir pres-sure and, because this is influenced by the compliance and buffering capacity of the aorta, future studies should directly measure the cyclic changes in aortic reservoir function and compliance.

PerspectivesTo our knowledge, this is the first study to examine hemo-dynamic mechanisms related to elevations in central BP with exercise. The results demonstrate that augmentation of cen-tral BP during exercise is mainly because of increases in for-ward propagating waves generated by left ventricular ejection. These incident waves are the principal components of excess pressure load induced with exercise and, in contrast to current theory, there seems to be only a minor role for wave reflec-tion in exercise central BP. These findings have relevance to understanding not only the physiology contributing to central BP waveform morphology, but also the pathophysiology of exercise hypertension.

Sources of FundingThis study was supported by a Grant-in-aid from the National Heart Foundation of Australia (reference G11H5915). M.G. Schultz received study funding from Exercise and Sports Science Australia (ESSA), Tom Penrose research grant, 2009; Dr Sharman was supported by a Career Development Award from the National Health and Medical Research Council of Australia (reference 569519). A.D. Hughes and Dr Davies received support from the UK National Institute for Health Research Biomedical Research Center Scheme and the British Heart Foundation Center of Research Excellence Award to Imperial College London. The other authors report no conflicts.

DisclosuresNone.

References 1. Schultz MG, Otahal P, Cleland VJ, Blizzard L, Marwick TH, Sharman JE.

Exercise-induced hypertension, cardiovascular events, and mortality in patients undergoing exercise stress testing: a systematic review and meta-analysis. Am J Hypertens. 2013;26:357–366.

2. Schultz MG, Gilroy D, Wright L, Bishop WL, Abhayaratna WP, Stowasser M, Sharman JE. Out-of-office and central blood pressure for risk stratification: a cross-sectional study in patients treated for hypertension. Eur J Clin Invest. 2012;42:393–401.

3. Nichols WW, O’Rourke MF. Mcdonald’s Blood Flow in Arteries: Theoretical, Experimental and Clinical Principles. London: Hodder Arnold; 2005.

4. Sharman J, Stowasser M, Fassett R, Marwick T, Franklin S. Central blood pressure measurement may improve risk stratification. J Hum Hypertens. 2008;22:838–844.

5. McEniery CM, Yasmin, McDonnell B, Munnery M, Wallace SM, Rowe CV, Cockcroft JR, Wilkinson IB; Anglo-Cardiff Collaborative Trial Investigators. Central pressure: variability and impact of cardiovascu-lar risk factors: the Anglo-Cardiff Collaborative Trial II. Hypertension. 2008;51:1476–1482.

6. Kroeker EJ, Wood EH. Comparison of simultaneously recorded central and peripheral arterial pressure pulses during rest, exercise and tilted posi-tion in man. Circ Res. 1955;3:623–632.

7. Rowell LB, Brengelmann GL, Blackmon JR, Bruce RA, Murray JA. Disparities between aortic and peripheral pulse pressures induced by upright exercise and vasomotor changes in man. Circulation. 1968;37:954–964.

8. Sharman JE, McEniery CM, Dhakam ZR, Coombes JS, Wilkinson IB, Cockcroft JR. Pulse pressure amplification during exercise is sig-nificantly reduced with age and hypercholesterolemia. J Hypertens. 2007;25:1249–1254.

9. Davies JE, Hadjiloizou N, Leibovich D, Malaweera A, Alastruey-Arimon J, Whinnett ZI, Manisty CH, Francis DP, Aguado-Sierra J, Foale RA, Malik IS, Parker KH, Mayet J, Hughes AD. Importance of the aortic res-ervoir in determining the shape of the arterial pressure waveform—the forgotten lessons of frank Artery Research. 2007;1:40–45.

10. Wang JJ, O’Brien AB, Shrive NG, Parker KH, Tyberg JV. Time-domain representation of ventricular-arterial coupling as a windkessel and wave system. Am J Physiol Heart Circ Physiol. 2003;284:H1358–H1368.

11. Davies JE, Baksi J, Francis DP, Hadjiloizou N, Whinnett ZI, Manisty CH, Aguado-Sierra J, Foale RA, Malik IS, Tyberg JV, Parker KH, Mayet J, Hughes AD. The arterial reservoir pressure increases with aging and is the major determinant of the aortic augmentation index. Am J Physiol Heart Circ Physiol. 2010;298:H580–H586.

12. Munir S, Jiang B, Guilcher A, Brett S, Redwood S, Marber M, Chowienczyk P. Exercise reduces arterial pressure augmentation through vasodilation of muscular arteries in humans. Am J Physiol Heart Circ Physiol. 2008;294:H1645–H1650.

13. Sharman JE, Lim R, Qasem AM, Coombes JS, Burgess MI, Franco J, Garrahy P, Wilkinson IB, Marwick TH. Validation of a generalized trans-fer function to noninvasively derive central blood pressure during exer-cise. Hypertension. 2006;47:1203–1208.

14. Davies JE, Whinnett ZI, Francis DP, Willson K, Foale RA, Malik IS, Hughes AD, Parker KH, Mayet J. Use of simultaneous pressure and velocity measurements to estimate arterial wave speed at a single site in humans. Am J Physiol Heart Circ Physiol. 2006;290:H878–H885.

15. Parker KH, Jones CJ. Forward and backward running waves in the arteries: analysis using the method of characteristics. J Biomech Eng. 1990;112:322–326.

16. Jones CJ, Sugawara M, Kondoh Y, Uchida K, Parker KH. Compression and expansion wavefront travel in canine ascending aortic flow: wave intensity analysis. Heart Vessels. 2002;16:91–98.

17. Hughes AD, Parker KH. Forward and backward waves in the arterial system: impedance or wave intensity analysis? Med Biol Eng Comput. 2009;47:207–210.

18. Baksi AJ, Treibel TA, Davies JE, Hadjiloizou N, Foale RA, Parker KH, Francis DP, Mayet J, Hughes AD. A meta-analysis of the mechanism of blood pressure change with aging. J Am Coll Cardiol. 2009;54:2087–2092.

19. Hope SA, Tay DB, Meredith IT, Cameron JD. Waveform dispersion, not reflection, may be the major determinant of aortic pressure wave morphol-ogy. Am J Physiol Heart Circ Physiol. 2005;289:H2497–H2502.

20. Davies JE, Alastruey J, Francis DP, Hadjiloizou N, Whinnett ZI, Manisty CH, Aguado-Sierra J, Willson K, Foale RA, Malik IS, Hughes AD, Parker KH, Mayet J. Attenuation of wave reflection by wave entrapment creates a “horizon effect” in the human aorta. Hypertension. 2012;60:778–785.




21. Klabunde R. Cardiovascular Physiolgy Concepts. Philadelphia: Lippincott Williams & Wilkins; 2005.

22. Sharman JE, McEniery CM, Campbell RI, Coombes JS, Wilkinson IB, Cockcroft JR. The effect of exercise on large artery haemodynamics in healthy young men. Eur J Clin Invest. 2005;35:738–744.

23. Murgo JP, Westerhof N, Giolma JP, Altobelli SA. Effects of exercise on aortic input impedance and pressure wave forms in normal humans. Circ Res. 1981;48:334–343.

24. Laskey WK, Kussmaul WG, Martin JL, Kleaveland JP, Hirshfeld JW Jr, Shroff S. Characteristics of vascular hydraulic load in patients with heart failure. Circulation. 1985;72:61–71.

25. Parker KH, Alastruey J, Stan GB. Arterial reservoir-excess pressure and ventricular work. Med Biol Eng Comput. 2012;50:419–424.

26. Mynard JP, Penny DJ, Davidson MR, Smolich JJ. The reservoir-wave par-adigm introduces error into arterial wave analysis: a computer modelling and in-vivo study. J Hypertens. 2012;30:734–743.

What Is New?•Central pressure and flow were measured during exercise to determine

the role of arterial wave travel on central blood pressure

•Central blood pressure rises during exercise because of increases in for-ward traveling waves generated by the heart.

What Is Relevant?• Although never proven, popular opinion suggests that rises in central blood

pressure during exercise are because of reflected waves from the periphery

of the body. Our findings provide evidence to the contrary, and represent an important advance in understanding central blood pressure physiology.

SummaryIncreased central blood pressure during exercise is caused mainly by forward propagating waves from the heart, whereas reflected waves play a minor role. These findings are significant with respect to understanding the physiology of central blood pressure and ex-ercise hypertension.

Novelty and Significance



1

Exercise central (aortic) blood pressure is predominantly driven

by forward travelling waves, not wave reflection.

Supplementary Material

Martin G. SCHULTZ MPhil, 1 Justin E. DAVIES MBBS, PhD,

2 Phillip ROBERTS-

THOMSON MBBS, 3

J. Andrew BLACK MBBS,3 Alun D. HUGHES MBBS, PhD,

2 and

James E. SHARMAN PhD.1

1 Menzies Research Institute Tasmania, University of Tasmania, Hobart, Tasmania,

AUSTRALIA. 2International Centre for Circulatory Health, National Heart & Lung Institute, Imperial

College London, UNITED KINGDOM. 3Royal Hobart Hospital, Hobart, Tasmania, AUSTRALIA.



2

Reproducibility of pressure and flow velocity measures. The within-subject standard

deviation of the difference between replicate recordings in the proximal aorta was ± 3.0

mmHg (coefficient of variation [CV], 6.5%) and ± 6.1 mmHg (CV, 8.2%) for aortic systolic

BP at rest and exercise respectively. For mean flow velocity, the standard deviation of the

mean difference was ± 2.5 ms-1

(CV, 15.4%) and ± 2.7 ms-1

(CV, 14.1%) at rest and exercise

respectively.

Wave intensity analysis. Utilising custom written software, the change in pressure was

separated into forward (dP+) and backward components (dP–),1 using equations 1 and 2

below, where was the density of blood (taken as 1050 kg.m–3

), and c was the wave speed

calculated using the single-point equation (equation 3), where dP was the incremental change

in pressure, and dU the incremental change in blood velocity.

Equation 1

Equation 2

Equation 3

Impedance analysis. Impedance analysis was performed as described by Hughes & Parker.1

The characteristic impedance (Z0) was calculated using a time-domain based approach1 and

separation of pressure into forward and backward components was undertaken using the

modification of the technique of Westerhof et al.,2 described by Laxminarayan.

3 This has

been shown to give nearly identical results to pressure separation using the wave intensity

approach.4

Reservoir and excess pressure separation. Aortic reservoir pressure (Preservoir) was

calculated with Equation 6 below, where P∞ is the pressure asymptote at which flow through

the microcirculation would be expected to cease, Pd is the diastolic pressure at t = 0, a is a

constant of proportionality, b = 1/RC where R is the resistance and C is the compliance of the

aortic reservoir.

Equation 4.



3

References

1. Hughes AD, Parker KH. Forward and backward waves in the arterial system:

Impedance or wave intensity analysis? Med Biol Eng Comput. 2009;47:207-210.

2. Westerhof N, Sipkema P, van den Bos GC, Elzinga G. Forward and backward waves

in the arterial system. Cardiovasc Res. 1972;6:648-656.

3. Laxminarayan S. The calculation of forward and backward waves in the arterial

system. Med Biol Eng Comput. 1979;17:130.

4. Hughes AD, Parker KH. Forward and backward waves in the arterial system:

Impedance or wave intensity analysis? Med Biol Eng Comput. 2009;47:207–210.



4

Figure S1. Example wave separation using time-domain (panels A and C) and frequency-

domain (panels B and D) analysis under resting (baseline) and exercise conditions. The black

line is the overall arterial pressure waveform, the blue line is the forward wave and the red

line is the backward wave.



5

Figure S2. Example impedance spectra plots for an individual at rest (baseline) and during

exercise.



exercise central (aortic) blood pressure is predominantly ... · we conclude that elevation of...

Documents